Can I use REGEX to remove last character in a file

[former_member201275]

I have a file on the sap server which looks something like this:

0001#Text#gfdkgh#

0002#more#ljjpoir#

0003#then#ljddpoiw#

I want a program to remove the last hash character from each line.

I can read this file with OPEN DATASET etc, then REVERSE the line, and REPLACE FIRST CHAR with SPACE. This works fine but I wanted to find out if it could be done with REGEX as it might be more economical?

All help greatly appreciated. Thank you.

[Former Member]

You could use

REPLACE FIRST OCCURRENCE OF REGEX '(.*)#$' IN line WITH '$1'.

but a simply

SHIFT line RIGHT DELETING TRAILING '#'.

should also do the job.

BR,

Gábor

[former_member201275]

ah yes! good idea! i will try this.

Thank you.

[Former Member]

Could it be that your server is UNIX and the file from Windows? I'd check the last character in debug - if it is x13 (CR), the file was likely FTPed in binary mode, not ASCII mode. Check the addition 'smart line feed' in 'open dataset'. It ignores CR.

I know, not a direct answer to your question, but it may save you some hassle...

Wolfgang

[former_member182550]

Well well well......

Mr Propfe - A very long time no see.  I hope you are doing well ?

Regards,

Rich

Oh - and to answer the question,  a # is not always a # - it's a non-printable character with an ascii code less than $0F.

Rich

[Clemenss]

Hi Glen,

regex is extremely powerful but quite slow. Test the perfromance difference of

REPLACE FIRST OCCURRENCE OF REGEX '(.*)#$' IN line WITH '$1'

against

SHIFT line RIGHT DELETING TRAILING CL_ABAP_CHAR_UTILITIES=>NEWLINE. "or CR_LF

The character displayed as '#' is probably a contraol character, check out the CL_ABAP_CHAR_UTILITIES attributes.

Regards  Clemens

.